Video background processing method, system and storage medium based on network signal strength
By monitoring and analyzing network signal data in real time, adjusting the ambiguity and dynamicity of video background processing, the lag and delay problems caused by network signal changes in online video communication are solved, and the user experience and video quality are improved.
Patent Information
- Application Number
- CN202411125439.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-08-16
AI Technical Summary
During online video communication, due to changes in network signal strength, the stability and fluency of video communication are easily reduced, and lag, delay or frame loss occurs, affecting the user experience.
By obtaining network signal data and communication screen images in real time, extracting signal intensity values and signal fluctuations, calculating network fluctuations, adjusting the blurring degree of static background images and the moving distance of dynamic screen images, and introducing edge smoothing values to optimize static backgrounds and dynamics The transition effect between the pictures, and dynamically adjust the number of interpolated frames and deformation degree according to network conditions to improve the smoothness and visual effect of video communication.
Effectively respond to network signal fluctuations, improve the user experience and video quality of online video communication, reduce lag and delay, and improve the overall quality and user satisfaction of video communication.
Smart Images

Figure CN118646930B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of video background processing, and in particular to a method, system and storage medium for video background processing based on network signal strength. Background Art
[0002] In current online video communication technology, a common problem is that the video communication process is often easily disturbed by the strength of the network signal. Due to the complex and changeable network environment, including bandwidth limitations, network congestion, signal attenuation and other factors, the network signal will change in real time. This change directly affects the stability and fluency of online video communication.
[0003] Specifically, when the network signal strength weakens or the network freezes, the transmission of video data will be affected, causing freezes, delays, or even frame losses in the screen and background of online video communications. This not only reduces the real-time and clarity of video communications, but also seriously affects the user's communication experience and satisfaction. Summary of the invention
[0004] In order to ensure the fluency of online video communication and alleviate the problem of video communication jamming, the present application provides a video background processing method, system and storage medium based on network signal strength.
[0005] In a first aspect, the present application provides a method for processing video background based on network signal strength, which adopts the following technical solution:
[0006] A video background processing method based on network signal strength comprises the following steps:
[0007] Based on instant video communication, network signal data and communication screen images are obtained in real time;
[0008] Extracting a signal strength value from the network signal data;
[0009] Calculate the signal fluctuation value of the signal strength value within the most recent first set time period;
[0010] If the signal fluctuation value is greater than a preset reference fluctuation value, a static background image and a dynamic screen image are extracted from the communication screen image;
[0011] Calculating a network fluency value according to the signal strength value and the signal fluctuation value;
[0012] If the network fluency value is higher than the preset reference fluency value, the blur degree of the static background image is adjusted according to the first difference between the network fluency value and the reference fluency value; blur degree = basic blur value + (first difference / reference fluency value maximum value × maximum blur increment); wherein the basic blur value, the reference fluency value maximum value and the maximum blur increment are preset values, and the reference fluency value maximum value is the maximum value of the network fluency;
[0013] The moving distance of the dynamic picture image is reduced according to the first difference; the moving distance reduction ratio=1-(first difference / reference smoothness value maximum value×scaling factor); wherein the scaling factor is a coefficient used to adjust the reduction ratio.
[0014] By adopting the above technical solution, during the online video communication, the network signal will change in real time. When the network freezes, the screen and background of the online video communication will also freeze. Therefore, based on the strength of the network signal, the edge transition of the screen and background is performed, or the background is identified, and the display characteristics of the screen and background are adaptively increased to compensate for the disadvantages of the screen and background freezing. This video background processing solution will be able to more effectively cope with network signal fluctuations and improve the user experience of online video communication.
[0015] Optionally, the method further comprises the following steps:
[0016] If a static background image and a dynamic screen image are extracted from the communication screen image, then in the most recent second set time period, the displacement of the static background image in the window is calculated as the background displacement speed value, and the relative displacement between the dynamic screen image and the static background image is calculated as the screen displacement speed value;
[0017] Calculating a dynamic speed value according to the background displacement speed and the picture displacement speed;
[0018] Calculating an edge smoothing value according to the dynamic speed value and the signal strength value;
[0019] The blur degree of the boundary area between the static background image and the dynamic picture image is adjusted according to the edge smoothing value.
[0020] By adopting the above technical solution, the processing of the dynamic characteristics of the video picture when the network signal changes is further refined, and the concept of edge smoothing is introduced to optimize the transition effect between the static background and the dynamic picture.
[0021] Optionally, the method further comprises the following steps:
[0022] Analyzing pixel changes between adjacent frames of the dynamic picture image, and predicting and generating intermediate frames to be inserted between original frames;
[0023] Adjusting the number of interpolated frames of the intermediate frames of the dynamic picture image according to the edge smoothing value;
[0024] The higher the edge smoothing value is, the more the number of interpolated frames is; the lower the edge smoothing value is, the fewer the number of interpolated frames is;
[0025] When the network fluency value is lower than the set value, reduce the number of interpolated frames by the set amount.
[0026] By adopting the above technical solution, the number of interpolated frames of dynamic picture images can be dynamically adjusted according to the edge smoothing value, which can further improve the fluency and viewing experience of online video communication, while taking into account the limitations of network conditions and device performance to achieve a better user experience.
[0027] Optionally, the method further comprises the following steps:
[0028] adjusting the deformation degree of the static background image according to the edge smoothing value;
[0029] The farther the static background image is from the dynamic picture image, the smaller the deformation degree of the static background image is, and the closer the static background image is to the dynamic picture image, the greater the deformation degree of the static background image is;
[0030] And the deformation direction of the static background image matches the movement direction of the dynamic picture image, wherein the angle between the vector of the deformation direction and the vector of the movement direction is an acute angle.
[0031] By adopting the above technical solution, the deformation degree of the static background image can be finely adjusted according to factors such as edge smoothing value, distance and movement direction, so as to further improve the visual effect and user experience of online video communication.
[0032] Optionally, the method further comprises the following steps:
[0033] The gain value of the deformation degree is adjusted according to the network strength value, wherein the lower the network strength value is, the smaller the gain value of the deformation degree is; and the higher the network strength value is, the larger the gain value of the deformation degree is;
[0034] And the size of the network strength value is positively correlated to adjust the size of the vector angle.
[0035] By adopting the above technical solutions, the network strength value reflects the stability and speed of the current network connection. In the case of poor network conditions, high computational complexity image processing operations, such as deformation of static background images, may cause video quality degradation due to bandwidth limitations or delays.
[0036] Optionally, the method further comprises the following steps:
[0037] Extracting the main outline of the graphics in the static background image;
[0038] According to the main outline of the graphic, a corresponding correction graphic is matched from a preset graphic library;
[0039] The correction graphic is placed between the static background image and the dynamic picture image, and the edges of the three are smoothed.
[0040] By adopting the above technical solution, the quality and efficiency of image synthesis can be significantly improved, and a more refined and natural image processing effect can be achieved.
[0041] Optionally, the calculation formula of the signal fluctuation value is as follows:
[0042] Signal fluctuation value = sqrt{(1 / N)×∑(xi-μ) 2}; where xi is the signal strength value at each time point in the first set time period, μ is the average value of these signal strength values, and N is the number of data points;
[0043] Based on multiple networked devices in the same local area network and the same instant video communication channel, signal fluctuation data of multiple devices are obtained;
[0044] Calculating the similarity between a plurality of signal fluctuation data, and adjusting the correction value of the correction graph according to the similarity;
[0045] The higher the similarity, the larger the correction value; the lower the similarity, the smaller the correction value.
[0046] By adopting the above technical solution, it is possible to systematically collect and analyze the signal fluctuation data of multiple devices in the same local area network, evaluate the similarity between them, and optimize the video communication quality or adjust the network settings based on the similarity.
[0047] Optionally, the similarity is calculated as follows:
[0048] Calculate signal fluctuation values respectively according to a plurality of signal fluctuation data;
[0049] Similarity = 1-standard deviation of signal fluctuation value / average value of signal fluctuation value × 100%.
[0050] By adopting the above technical solution, the standard deviation of the signal fluctuation value / the average value of the signal fluctuation value represents the degree of discreteness of the data. If the degree of discreteness is higher, it means that the similarity is lower.
[0051] In a second aspect, the present application provides a video background processing system based on network signal strength, which adopts the following technical solution:
[0052] A video background processing system based on network signal strength comprises a processor, wherein the program of any one of the above-mentioned video background processing methods based on network signal strength runs on the processor.
[0053] In a third aspect, the present application provides a storage medium, which adopts the following technical solution:
[0054] A storage medium stores a program of any one of the above-mentioned methods for processing video background based on network signal strength.
[0055] In summary, the present application includes at least one of the following beneficial technical effects:
[0056] Improve video communication quality: By real-time monitoring of the network signal strength of each device during video communication and calculating its fluctuation value, the system can promptly detect and respond to changes in network conditions. When the signal fluctuates greatly, corresponding adjustment measures can be taken to reduce video freezes, delays or packet loss caused by network instability, thereby improving the overall quality of video communication.
[0057] Optimize network resource allocation: Based on the signal fluctuation data of multiple devices, the system can evaluate the allocation of network resources in the entire LAN. When it is found that the signal fluctuation of some devices is large due to network congestion, the network resource allocation strategy can be dynamically adjusted to provide more bandwidth or priority for these devices to ensure the smoothness of video communication. At the same time, it can also avoid unnecessary waste of resources and improve the overall utilization efficiency of network resources.
[0058] Enhanced user experience: By reducing problems such as freezes and delays in video communications, users can get a smoother and more natural video communication experience. In addition, the system can automatically adjust the display effect of the video interface according to changes in network signal strength to adapt to different network environments, further improving the user's viewing experience.
[0059] Intelligent fault prediction and prevention: Through long-term monitoring and analysis of signal fluctuation data, the system can identify early signs of network failure or performance bottlenecks. Once potential problems are discovered, the system can take preventive measures in advance to avoid network failures from having a serious impact on video communications. This intelligent fault prediction and prevention capability helps improve the stability and reliability of the system.
[0060] Supporting multi-device collaboration: In the same LAN, different devices may have different network signal strengths and performance characteristics. The processing method based on signal fluctuation data can support the collaboration between multiple devices. By optimizing resource allocation and adjusting communication strategies, it ensures that each device can maintain a relatively consistent performance level during video communication, thereby improving the collaborative efficiency and stability of the entire system. BRIEF DESCRIPTION OF THE DRAWINGS
[0061] Figure 1 The invention is a step diagram of a method for processing a video background based on network signal strength.
[0062] Figure 2 It is a step diagram that introduces the concept of edge smoothing to optimize the transition between static background and dynamic picture.
[0063] Figure 3 It is a step diagram for dynamically adjusting the number of interpolated frames of a dynamic picture image according to an edge smoothing value. DETAILED DESCRIPTION
[0064] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings.
[0065] In the description of this specification, the description with reference to the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.
[0066] The present application embodiment discloses a video background processing method based on network signal strength, referring to Figure 1 , including the following steps:
[0067] Based on instant video communication, network signal data and communication screen images are obtained in real time;
[0068] Extract signal strength value from network signal data;
[0069] The signal fluctuation value of the signal strength value within the most recent first set time period is calculated; the signal fluctuation value can reflect the current network status.
[0070] If the signal fluctuation value is greater than the preset reference fluctuation value, the static background image and the dynamic picture image are extracted from the communication picture image. The signal fluctuation value is used to determine whether to separate the images to reduce unnecessary computational overhead. Image segmentation can use image processing techniques such as background difference, frame difference, and deep learning models to effectively distinguish between static background images and dynamic picture images.
[0071] The network fluency value is calculated based on the signal strength value and the signal fluctuation value to more comprehensively reflect the network quality.
[0072] If the network fluency value is higher than the preset reference fluency value, the blurriness of the static background image is adjusted according to the first difference between the network fluency value and the reference fluency value. Blur degree = basic blur value + (first difference / maximum value of reference fluency value × maximum blur increment); that is, the larger the first difference, the blurrier the static background image; the smaller the first difference, the clearer the static background image. Among them, the basic blur value, the maximum value of the reference fluency value and the maximum blur increment are preset values, and the maximum value of the reference fluency value is the maximum value of the network fluency. By dynamically adjusting the blurriness of the static background, the visual discomfort of the user can be reduced when the network is not good.
[0073] The moving distance of the dynamic picture image is reduced according to the first difference; the moving distance reduction ratio = 1-(first difference / maximum value of reference smoothness value × scaling factor); wherein the scaling factor is a coefficient used to adjust the reduction ratio. That is, the larger the first difference, the larger the reduction factor of the moving distance of the dynamic picture image; the smaller the first difference, the smaller the reduction factor of the moving distance of the dynamic picture image. Reducing the moving distance of the dynamic picture image can reduce the jumping feeling of the picture when the network is stuck, making the video smoother.
[0074] During online video communication, the network signal will change in real time. When the network freezes, the screen and background of online video communication will also freeze. Therefore, based on the strength of the network signal, the edge transition of the screen and background is performed, or the background is identified, and the display characteristics of the screen and background are adaptively increased to compensate for the shortcomings of the screen and background freeze. This video background processing solution will be able to more effectively cope with network signal fluctuations and improve the user experience of online video communication.
[0075] Reference Figure 2 , the method further comprises the steps of:
[0076] If a static background image and a dynamic screen image are extracted from the communication screen image, the displacement of the static background image in the window is calculated as the background displacement speed value in the most recent second set time period. The background displacement speed value reflects the overall movement speed of the background image. The system first separates the static background image and the dynamic screen image from the continuous video frames through algorithms such as background difference method and Gaussian mixture model; static background images, such as the fixed scene at the entrance of the park; dynamic screen images, such as walking pedestrians and passing vehicles. Assume that the most recent second set time period is the past 5 seconds. During these 5 seconds, the system detected that the static background image had a slight displacement in the window due to natural factors such as slight movement of the camera or wind blowing trees. By comparing the background images before and after these 5 seconds, the average displacement of the background image is calculated, which is the background displacement speed value. This value reflects the overall movement speed of the background image.
[0077] The relative displacement between the dynamic image and the static background image is calculated as the image displacement speed value, which reflects the movement speed of the dynamic object in the video image. For dynamic images, such as pedestrians, the system calculates the relative displacement between them and the static background image in the same time period. For example, when a pedestrian walks from one side of the screen to the other side, the relative displacement between the pedestrian and the background image is calculated, which is the image displacement speed value, reflecting the movement speed of the dynamic object in the video screen.
[0078] The dynamic speed value is calculated based on the background displacement speed and the screen displacement speed; the dynamic speed value can be obtained by weighted average calculation method, which reflects the overall dynamics of the video screen. For example, if the background displacement is small but the objects in the screen move quickly, the dynamic speed value will be biased towards the screen displacement speed value.
[0079] The edge smoothing value is calculated based on the dynamic speed value and the signal strength value; the edge smoothing value is calculated based on the dynamic speed value and the signal strength value. When the dynamic speed value is high, it means that there are more moving elements in the picture, and edge smoothing needs to be increased to reduce the visual impact caused by the movement; at the same time, when the signal strength value is low, edge smoothing tends to be increased to compensate for the unsmooth feeling caused by network jams.
[0080] The blur level of the boundary area between the static background image and the dynamic screen image is adjusted according to the edge smoothing value. If the edge smoothing value is larger, the blur level of the boundary area will be higher, thus achieving a smoother transition effect; otherwise, the blur level will be lower, maintaining the clarity of the image.
[0081] The processing of the dynamic characteristics of the video image when the network signal changes has been further refined, and the concept of edge smoothing has been introduced to optimize the transition effect between the static background and the dynamic image.
[0082] Reference Figure 3 , the method further comprises the steps of:
[0083] Analyze the pixel changes between adjacent frames of dynamic images and predict the generation of intermediate frames to be inserted between the original frames; through frame rate upscaling (FRUS) or motion interpolation technology, analyze the pixel differences (i.e. motion vectors) between consecutive frames, the algorithm can predict the intermediate images that should exist between these frames and generate these intermediate frames (also called interpolation frames). These interpolation frames are then inserted between the original frames, thereby increasing the frame rate without changing the video content, making the video look smoother.
[0084] The number of interpolated frames for the middle frames of the dynamic picture image is adjusted according to the edge smoothing value; the edge smoothing value is a measure of the edge clarity in the image. In video processing, the clarity of edges, such as the boundaries of objects, is crucial to the overall visual quality. If the edges are too blurry, not many interpolated frames are needed to avoid introducing unnecessary blur; on the contrary, if the edges are clear, more interpolated frames are needed to better capture the motion details. Therefore, adjusting the number of interpolated frames according to the edge smoothing value is an optimization strategy that aims to maintain the clarity of the video while improving the smoothness.
[0085] The higher the smoothing value, the more interpolated frames are; the lower the smoothing value, the fewer interpolated frames are. A high smoothing value usually means that the image is blurry, and increasing the number of interpolated frames can help fill in details and make the motion look smoother. On the other hand, a low smoothing value indicates that the image has clear edges, and too many interpolated frames may introduce unnecessary blur or distortion, so the number of interpolated frames should be reduced.
[0086] When the network fluency value is lower than the set value, reduce the number of interpolated frames; this step takes into account the actual situation of network transmission. In streaming media playback, the quality of the network directly affects the smoothness of video playback. If the network fluency value is lower than the set value, it means that the current network environment is not good, and continuing to play at a high frame rate may cause buffering or jamming. Therefore, reducing the number of interpolated frames can reduce the demand for bandwidth and reduce transmission pressure, thereby improving the stability of video playback to a certain extent.
[0087] Dynamically adjusting the number of interpolated frames of dynamic images according to the edge smoothing value can further improve the fluency and viewing experience of online video communication, while taking into account the limitations of network conditions and device performance to achieve a better user experience.
[0088] In order to further improve the visual effect and user experience of online video communication, the method also includes the following steps:
[0089] The deformation degree of the static background image is adjusted according to the edge smoothing value; deformation is the deformation of different parts to different degrees. The edge smoothing value is not only used to adjust the number of interpolated frames of the dynamic picture image, but also to control the deformation degree of the static background image. Here, the edge smoothing value is used as a comprehensive indicator to reflect the overall clarity and sharpness of the image. A lower edge smoothing value means a clearer image and allows a smaller deformation degree to maintain the authenticity of the background; while a higher edge smoothing value means a blurrier image and allows a larger deformation to simulate a more natural dynamic effect.
[0090] The farther the static background image is from the dynamic picture image, the smaller the deformation of the static background image is, and the closer the static background image is to the dynamic picture image, the greater the deformation of the static background image is. Based on the principle of visual perception, that is, the visual changes of distant objects are relatively small, while the changes of nearby objects are more obvious. Therefore, when the static background image is far away from the dynamic picture image, such as a moving object or person, its deformation degree should be reduced accordingly to maintain the stability of the background; and when the background image is close to the dynamic picture, the deformation degree should be increased to simulate the background changes caused by the movement of the dynamic object.
[0091] The deformation direction of the static background image matches the movement direction of the dynamic picture image, wherein the angle between the deformation direction vector and the movement direction vector is an acute angle. This ensures the coordination and consistency between the background deformation and the dynamic picture. The acute angle between the deformation direction vector and the movement direction vector means that the trend of the background deformation should be consistent with the movement direction of the dynamic picture or present a certain auxiliary effect. This helps to enhance the immersion and realism of the video, making it easier for the audience to accept and integrate into the scene presented by the video.
[0092] Fine-tune the degree of deformation of static background images based on factors such as edge smoothing value, distance and direction of movement.
[0093] In order to help maintain the overall coherence and watchability of the video picture, the method also includes the following steps:
[0094] The gain value of the deformation degree is adjusted according to the network strength value. The lower the network strength value, the smaller the gain value of the deformation degree; the higher the network strength value, the larger the gain value of the deformation degree. When the network strength value is low, it means that the network transmission conditions are not good, such as delays, packet loss and other problems. In this case, reducing the gain value of the deformation degree can reduce the consumption of video processing resources and reduce the risk of video quality degradation due to network fluctuations. Doing so helps maintain the basic smoothness of the video. On the contrary, when the network strength value is high, the network transmission conditions are good and the video processing resources are relatively abundant. At this time, increasing the gain value of the deformation degree can further enhance the visual effect of the video, making the interaction between the dynamic picture and the static background more natural and realistic.
[0095] The network strength value is positively correlated with the size of the vector angle. The network strength value not only affects the gain value of the deformation degree, but also directly affects the size of the angle between the deformation direction and the movement direction vector. When the network strength value is high, the angle between the deformation direction and the movement direction is allowed to increase, which means that the deformation can more freely follow or assist the movement of the dynamic picture, creating a more vivid and rich visual effect. However, when the network strength value is low, the size of the angle between the deformation direction and the movement direction vector should be reduced to ensure that the deformation does not appear abrupt or unstable due to network fluctuations.
[0096] In order to improve the quality and efficiency of image synthesis and achieve more refined and natural image processing, the method further includes the following steps:
[0097] Extract the main outline of the graphics in the static background image; through image processing technology, such as edge detection, contour extraction and other algorithms, identify and extract the outline of the main graphics from the static background image. These outlines represent the most important visual elements in the image and provide a basis for subsequent graphics matching and correction.
[0098] According to the main outline of the graphic, the corresponding correction graphic is matched from the preset graphic library; after the main outline of the graphic is extracted, the system will use these outlines as query conditions to search and match in the preset graphic library. The graphic library should contain a variety of predefined graphic templates, which can be standard geometric shapes, common object outlines, or specific graphics for specific application scenarios. By comparing the shape, size, proportion and other features of the outline, the system can find the correction graphic that best matches the extracted outline.
[0099] Place the correction graphic between the static background image and the dynamic picture image, and smooth the edges of the three. After finding the matching correction graphic, the next step is to cleverly integrate it into the video frame. That is, place the correction graphic in the appropriate position between the static background image and the dynamic picture image to form visual coherence and harmony. Edge smoothing is the key in this step. Since the introduction of the correction graphic may produce obvious boundaries or seams between images, these seams may distract the audience and reduce the viewing experience of the video. Therefore, it is necessary to smooth the edges between the correction graphic and the static background and dynamic picture to reduce or eliminate the visibility of the seams. This can be achieved through image fusion techniques such as feathering and gradient, so that the correction graphic is seamlessly connected with the surrounding images.
[0100] The calculation formula of the signal fluctuation value is as follows:
[0101] Signal fluctuation value = sqrt{(1 / N)×∑(xi-μ) 2}; where xi is the signal strength value at each time point in the first set time period, μ is the average of these signal strength values, and N is the number of data points. The signal fluctuation value is essentially obtained through the calculation formula of the standard deviation, which is used to measure the discrete degree or fluctuation size of the signal strength.
[0102] Based on multiple networked devices in the same local area network and the same instant video communication channel, signal fluctuation data of multiple devices are obtained.
[0103] The similarity between multiple signal fluctuation data is calculated, and the correction value of the correction graph is adjusted according to the similarity; the similarity calculation method can be based on a variety of statistical metrics, such as correlation coefficient, distance metrics, such as Euclidean distance, Manhattan distance, etc. Devices with high similarity mean that the signal fluctuation patterns are similar and are affected by similar network conditions.
[0104] The higher the similarity, the larger the correction value; the lower the similarity, the smaller the correction value. A higher similarity means that multiple devices are running under the same network conditions. In this case, the correction value can be appropriately increased to enhance the visual effect and stability of the video. On the contrary, a lower similarity may mean that the network conditions faced by different devices are quite different. In this case, the correction value should be reduced to avoid video distortion or instability caused by over-correction.
[0105] It can systematically collect and analyze signal fluctuation data of multiple devices in the same LAN, evaluate the similarity between them, and optimize video communication quality or adjust network settings based on the similarity.
[0106] The similarity calculation formula is as follows:
[0107] The signal fluctuation values are calculated respectively according to the plurality of signal fluctuation data; the calculation formula of the signal fluctuation value is the same as the above-mentioned calculation formula of the signal fluctuation value.
[0108] Similarity = 1-standard deviation of signal fluctuation value / average value of signal fluctuation value × 100%. When the standard deviation is smaller than the average value, the similarity is close to 1, indicating that the signal fluctuations of different devices are relatively consistent; when the standard deviation is larger than the average value, the similarity decreases, indicating that the signal fluctuations are more different.
[0109] The standard deviation of the signal fluctuation value / the average value of the signal fluctuation value represents the degree of dispersion of the data. The higher the dispersion, the lower the similarity.
[0110] For example, suppose the signal fluctuation values of multiple devices are: 0.5, 0.6, 0.55, 0.58, 0.62;
[0111] The calculated standard deviation is 0.0344 and the mean is 0.57;
[0112] Calculated similarity = 1-6.04% = 93.96%.
[0113] An embodiment of the present application further discloses a video background processing system based on network signal strength, including a processor, wherein a program of any one of the above-mentioned video background processing methods based on network signal strength is running on the processor.
[0114] The embodiment of the present application further discloses a storage medium storing a program of any one of the above-mentioned methods for processing video background based on network signal strength.
[0115] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.
Claims
1. A video background processing method based on network signal strength, characterized in that: The steps include: Based on instant video communication, network signal data and communication screen images are obtained in real time; Extracting a signal strength value from the network signal data; Calculate the signal fluctuation value of the signal strength value within the most recent first set time period; If the signal fluctuation value is greater than a preset reference fluctuation value, a static background image and a dynamic screen image are extracted from the communication screen image; Calculating a network fluency value according to the signal strength value and the signal fluctuation value; If the network fluency value is higher than a preset reference fluency value, adjusting the blur level of the static background image according to a first difference between the network fluency value and the reference fluency value; Fuzziness degree = basic fuzziness value + (first difference / reference fluency value maximum value × maximum fuzziness increment); wherein the basic fuzziness value, the reference fluency value maximum value and the maximum fuzziness increment are preset values, and the reference fluency value maximum value is the maximum value of the network fluency; Reducing the moving distance of the dynamic picture image according to the first difference; the moving distance reduction ratio=1-(first difference / reference smoothness value maximum value×scaling factor); wherein the scaling factor is a coefficient used to adjust the reduction ratio; If a static background image and a dynamic screen image are extracted from the communication screen image, then in the most recent second set time period, the displacement of the static background image in the window is calculated as the background displacement speed value, and the relative displacement between the dynamic screen image and the static background image is calculated as the screen displacement speed value; Calculating a dynamic speed value according to the background displacement speed and the picture displacement speed; Calculating an edge smoothing value according to the dynamic speed value and the signal strength value; The blur degree of the boundary area between the static background image and the dynamic picture image is adjusted according to the edge smoothing value.
2. The video background processing method based on network signal strength according to claim 1 is characterized in that: The method further comprises the steps of: Analyzing pixel changes between adjacent frames of the dynamic picture image, and predicting and generating intermediate frames to be inserted between original frames; Adjusting the number of interpolated frames of the intermediate frames of the dynamic picture image according to the edge smoothing value; The higher the edge smoothing value is, the more the number of interpolated frames is; the lower the edge smoothing value is, the fewer the number of interpolated frames is; When the network fluency value is lower than the set value, reduce the number of interpolated frames by the set amount.
3. The video background processing method based on network signal strength according to claim 2 is characterized in that: The method further comprises the steps of: adjusting the deformation degree of the static background image according to the edge smoothing value; The farther the static background image is from the dynamic picture image, the smaller the deformation degree of the static background image is, and the closer the static background image is to the dynamic picture image, the greater the deformation degree of the static background image is; And the deformation direction of the static background image matches the movement direction of the dynamic picture image, wherein the angle between the vector of the deformation direction and the vector of the movement direction is an acute angle.
4. The video background processing method based on network signal strength according to claim 3 is characterized in that: The method further comprises the steps of: The gain value of the deformation degree is adjusted according to the network strength value, wherein the lower the network strength value is, the smaller the gain value of the deformation degree is; and the higher the network strength value is, the larger the gain value of the deformation degree is; And the size of the network strength value is positively correlated to adjust the size of the vector angle.
5. The video background processing method based on network signal strength according to claim 1 is characterized in that: The method further comprises the steps of: Extracting the main outline of the graphics in the static background image; According to the main outline of the graphic, a corresponding correction graphic is matched from a preset graphic library; The correction graphic is placed between the static background image and the dynamic picture image, and the edges of the three are smoothed.
6. The video background processing method based on network signal strength according to claim 5 is characterized in that: The calculation formula of the signal fluctuation value is as follows: Signal Fluctuation Value = ; Wherein, xi is the signal strength value at each time point in the first set time period, μ is the average value of these signal strength values, N is the number of data points; sqrt{} is the square root function; ∑() is the summation function; Based on multiple networked devices in the same local area network and the same instant video communication channel, signal fluctuation data of multiple devices are obtained; Calculating the similarity between a plurality of signal fluctuation data, and adjusting the correction value of the correction graph according to the similarity; The higher the similarity, the larger the correction value; the lower the similarity, the smaller the correction value.
7. The video background processing method based on network signal strength according to claim 6 is characterized in that: The calculation formula of the similarity is as follows: Calculate signal fluctuation values respectively according to a plurality of signal fluctuation data; Similarity = 1-standard deviation of signal fluctuation value / average value of signal fluctuation value × 100%.
8. A video background processing system based on network signal strength, characterized in that: It comprises a processor, wherein the processor executes the steps of the video background processing method based on network signal strength as described in any one of claims 1 to 7.
9. A storage medium storing a program, wherein the program, when executed by a processor, implements the steps of the video background processing method based on network signal strength as described in any one of claims 1 to 7.
